Tech moves fast! Stay ahead of the curve with Techopedia!
Join nearly 200,000 subscribers who receive actionable tech insights from Techopedia.
Ground truth is a term used in statistics and machine learning that means checking the results of machine learning for accuracy against the real world. The term is borrowed from meteorology, where "ground truth" refers to information obtained on site. The term implies a kind of reality check for machine learning algorithms.
"Ground truth" is a term borrowed from meteorology for independent confirmation at a site for information obtained by remote sensing. One example is a storm spotter reporting a tornado out in the field that a meteorologist is tracking on Doppler radar.
Researchers in machine learning also employ a kind of "ground truth," checking the classifications that machine learning algorithms make against what they know in reality. An example of ground truthing in this context is checking the messages a Bayesian spam filter has marked as spam for false positives, or email messages that are legitimate.
These kinds of tests allow researchers to refine their algorithms to make them more accurate.